DESCRIPTION (provided by candidate): Color vision requires a comparison of outputs from different types of color photoreceptor cells. In vertebrates, the initial steps of this processing are performed by retinal neurons that connect to the color photoreceptors. In fruit flies, such connections are likely to occur in the medulla, a brain structure close to the fly retina. In principle, a number of different comparisons could be made between four different types of fly color photoreceptors. We plan to identify neurons which contact two or more different types of photoreceptors. This, together with our knowledge of photoreceptor properties, will provide a better understanding of the mechanisms of color processing in flies. We expect that the basic principles that we will uncover will apply to color vision in different animals, including humans. To this end, we propose to develop three different approaches for visualizing neurons connected to the photoreceptors: by randomly labeling different sets of neurons, by identifying markers for specific sets of medulla neurons and by directly labeling neurons connected to the color photoreceptors. Our work will provide a solid foundation for further studies involving genetic manipulations of color-processing neurons and further understanding of color perception .